Interaction between Face Detection and Learning Tracking Systems for Autonomous Robots

Domestic service robots are becoming increasingly available and affordable. However, interaction with the real world is still a challenge. Fast and reliable visual perception is one of these challenges. Systems that track the target over time are fast and can yield good results because they exploit temporal coherence, but they often need manual initialization. In this study, such a tracking system is combined with a detector system that initializes and corrects it automatically. Different combinations of features are explored and evaluated on a dataset. The combined detection and tracking system is used to guide a mobile, autonomous robot in the real world. Real-world data is used to continuously train the tracker system, to make it more robust to changes in illumination conditions and other environmental conditions. This robustness is necessary for the tracking system to handle the changing conditions in the real world with only limited feedback from